INDEX

Explanations

numerical data related to statistical analysis and comparisons

oai_token-act-pair · gpt-4o-mini Triggered by @bot

ranges or categories

np_acts-logits-general · gemini-2.0-flash

numbers, ranges, binary, categories

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 שוליים

-0.58

 propOrder

-0.57

potranspiration

-0.55

quelize

-0.54

ponses

-0.54

ьаж

-0.54

OGND

-0.51

Assured

-0.50

enderror

-0.50

GOTREF

-0.49

POSITIVE LOGITS

 betweenstory

0.52

 ranges

0.42

 range

0.39

AllowUser

0.38

="@+

0.36

 near

0.35

 ordered

0.35

uyan

0.35

Ranges

0.34

 Range

0.33

Activations Density 2.682%

numerical data related to statistical analysis and comparisons

ranges or categories

numbers, ranges, binary, categories

No Comments

No Known Activations